A healthy unmarried woman was undergoing a dilation and curettage (D&C) following an incomplete spontaneous abortion (miscarriage).

At this community hospital, a new operating room (OR) suite had recently opened. It was equipped with video cameras in all OR rooms to check staff location and activities, observe the status of ongoing procedures, and assist with development of educational materials. The video cameras are monitored at the nurses' station, located just inside the OR suite. As such, the monitors are visible to anyone who enters the OR doors, and sometimes to those standing outside the doors. Prior to surgery, the hospital admissions staff obtains general consent from patients for videotaping for "education and safety" purposes. The use, placement, and high visibility of the cameras in the OR is not explained in writing or discussed with the patient.

During the D&C procedure, the woman's face was shown on the OR video monitor. She was recognized by someone who passed the OR suite when the door was open. The serious privacy issues came to light after the passerby disclosed the woman's presence in the OR to other people. "Gossip" spread around town about the woman's pregnancy and D&C. It raised a great deal of speculation and was embarrassing to the woman and others.

by Colin F. Mackenzie, MD

The sequence of events described in this case—facial recognition of a patient in a community hospital undergoing a surgical procedure—can occur for several reasons unrelated to the use of video in the operating room. The patient may be recognized in the same-day surgery admission center, while being transported on a gurney or wheelchair, or during entry into the OR itself. The Health Insurance Portability and Accountability Act (HIPAA) regulations do not protect privacy for these chance encounters, since the procedure to be performed is listed on the OR schedule and, even with no names identified, this information can be linked to the OR room number that the patient entered.

In the past few years, use of video in hospitals nationwide has increased because of security concerns and because video equipment is inexpensive, unobtrusive, and high quality images are easily obtained. Besides security surveillance, video images are also being used for research, educational, and quality management purposes in the pre-hospital, emergency department, outpatient clinic, and operating room during everyday patient interactions with clinicians.(1) However, the example in this case shows how video can unintentionally invade the privacy of patients and result in revelation of confidential information. Such breaches of privacy can have long-lasting impact: one case from several years ago, in which the video record of unsuccessful resuscitation was inadvertently displayed to the patient's recently bereaved family members in a waiting room, continues to be discussed in a variety of fora. This commentary will identify regulatory, technical, and procedural means to approach privacy and confidentiality issues of video in the hospital workspace.

Implemented April 16, 2003, the HIPAA Privacy Rule [45 UFR Prts 160 and 164] covers protected health information (PHI), "about individuals' identifiable health information, transmitted or maintained in any form or medium." The Privacy Rule requires "patient authorization for use or disclosure of PHI," but there are exceptions such as "to avert serious threat to health or patient safety." Moreover, collection of limited data with "indirect identifiers such as age, dates of service, zip codes" is allowed as long as there are no direct identifiers (eg, names, social security numbers). Another exception is if there is a waiver review by the Institutional Review Board (IRB) of the Privacy Rule for Patient Authorization. This IRB waiver on use or disclosure of PHI is only likely to be granted if it involves no more than minimal risk to the privacy of individuals.

Although most people think of names or social security numbers when considering PHI, HIPAA regulations specifically include finger and voice prints, full face photographic images, and any comparable images as types of PHI. In this case, the privacy issue was a full-face photographic image-quality video of the patient's face seen by a passerby; the case does not state whether audio was also used. HIPAA regulations state that this PHI, allowing facial recognition on the video, should not be used without patient authorization.(2) Perhaps of equal relevance to this case are the Joint Commission on Accreditation of Healthcare Organizations' (JCAHO) regulations about video recording in hospitals. These state that using video cameras in hospitals is permissible, provided that appropriate signs are posted around the hospital locations where cameras are in use. The wording suggested for these signs to be in compliance with these JCAHO regulations is "Be aware that filming is in use" or "Video taping underway."(3) There are clearly fewer risks of invasion of patient privacy with video surveillance (as in this case) compared to video recording, because video surveillance is only a remote form of observation with no permanent record of the event.(4)

Video observation is one mean of transmitting what is happening in a given OR to a central location or workstation for OR scheduling, monitoring, and coordination. Video provides useful work-related knowledge to OR workers, such as cleaning staff, surgeons, anesthesiologists, nurses, technicians, and patient transport personnel.(5-7) The video images can also help staff track OR start and finish times. The question therefore becomes, how can this technology's organizational usefulness be maintained while avoiding privacy infractions and preserving confidentiality?

Our solution to this conundrum—how to take advantage of the usefulness of video without compromising confidentiality—is to degrade the images (Video 1), avoid the use of sound, and display the images only in a restricted area of the OR. Higher quality images never remain on display for longer than 2 minutes, and can be displayed (in restricted areas only, such as in zones restricted to providers wearing scrubs, and the main OR scheduling office) only after an authorized access card is swiped. OR images used for organizational and logistic purposes (ie, to determine whether cases have begun or ended) are not recorded.(6-7) Where photographic-quality video images are necessary, we use camera angles and image border controls that allow us to avoid recognition of individuals; we also blur patients' facial features (Video 2). Finally, we remove patient identifiers from paperwork associated with video records, and strictly limit access to these records: they are stored under lock and key, only for as long as required for analysis before their destruction by degaussing.(4)

Studies using video have been carried out in the medical domain since the 1950s.(8) The recent availability of inexpensive high-quality video cameras and recorders has resulted in a huge increase in the use of video recording in many different medical settings, including to evaluate medical student and physician clinical performance (9), critique physical exams of patients, and assess interpersonal skills of surgeons.(10) Video has also been used for ergonomic analysis of the clinician's workplace (11) as well as to determine conformity to patient management protocols such as Advanced Trauma and Cardiac Life Support.(12) Quality management and research in human factors are among other reasons for use of video recording in the clinical domain.(13) These varied uses of video recording are likely to increase over time, making it critical to build and maintain robust systems of preserving patient (and, where appropriate, provider) confidentiality.

Take-Home Points

• Respect privacy and display confidential information only to authorized personnel.
• Use degraded images where public viewing is possible.
• Avoid the use of sound.
• Display the video images so that they are shown only in restricted areas of the operating room.
• Limit access to remote video image review using authorization card swipe access and restricted viewing periods.

Colin F. Mackenzie, MD
Professor and Director
National Study Center for Trauma & EMS
University of Maryland

1. Mackenzie CF, Hu P F-M, Xiao Y, Seagull JF. Video acquisition and audio system network (VAASNET®) for analysis of workplace safety performance. Biomed Inst Tech. 2003;37:285-91.

2. United States Department of Health and Human Services, Office for Civil Rights. Medical Privacy: National Standards to Protect the Privacy of Personal Health Information.
[ go to related site ]. Accessed March 5, 2004.

3. Joint Commission on Accreditation of Healthcare Organizations. Patients Rights and informed consent when videotaping or filming.
[ go to related site ]. Accessed March 5, 2004.

4. Mackenzie CF, Xiao Y. Video techniques and data compared with observation in emergency trauma care. Qual Saf Health Care. 2003;12 Suppl 2:ii51-7.
[ go to pubmed ]

5. Xiao Y, Lasome C, Moss, et al. Cognitive properties of a whiteboard: A case study in a trauma center. In Printz W, Jarke M, Rogers Y, Schmidt K, Wulf V, eds. Proceedings of the seventh european conference on computer supported cooperative work. 16-20 September 2001. Bonn, Germany. Kluwer Academic Publishers; 2001:259-278.

6. Xiao Y, Hu P F-M, Seagull JF, Mackenzie CF. Distributed planning and monitoring in a dynamic environment:, trade-offs of information access and privacy. 2003 proceedings of IEEE international conference on systems man, and cybernetics; 2003: 4141-46.
[ go to related site ]. Accessed February 20, 2004.

7. Xiao, Y, Seagull JF, Hu, P F-M, Mackenzie CF, Gilbert TB. Distributed monitoring and a video-based toolset. 2003 proceedings of IEEE international conference on systems, man, and cybernetics; 2003: 1778-83. Available at:
[ go to related site ]. Accessed February 20, 2004.

8. Xiao Y, Mackenzie CF. Stress and decision making in trauma patient resuscitation. Final report available at:
[ go to related site ]. Accessed March 5, 2004.

9. Tardiff K. A videotape technique for measuring clinical skills: three years of experience. J Med Educ. 1981;56:187-91.
[ go to pubmed ]

10. Burchard KW, Rowland-Morin PA. A new method of assessing the interpersonal skills of surgeons. Acad Med. 1990;65:274-6.
[ go to pubmed ]

11. Harper BD, Mackenzie CF, Norman KL. Qualitative measures in the ergonomic examination of the trauma resuscitation unit's anesthesia workspace. Proc Hum Fact and Ergonomics Soc, 47th meeting. 1995;2:723-7.

12. Hoyt DB, Shackford SR, Fridland PH, et al. Video recording trauma resuscitations: an effective teaching technique. J Trauma. 1988;28:435-40.
[ go to pubmed ]

13. Video as Research Data Conference. National Study Center for Trauma and Emergency Medical Systems Web site. Available at
[ go to related site ]. Accessed January 11, 2004.